bjh
research paper
Haptoglobin, alpha-thalassaemia and glucose-6-phosphate
dehydrogenase polymorphisms and risk of abnormal
transcranial Doppler among patients with sickle cell
anaemia in Tanzania
Sharon E. Cox, 1 ' 2 Julie Makani, 2 ' 3
Deogratias Soka, 2 Veline S.
L'Esperence, 4 Edward Kija, 2 Paula
Dominguez-Salas, 1 Charles R. J.
Newton, 2 ' 5 ' 6 Anthony A. Birch, 7 Andrew
M. Prentice 1 and Fenella J. Kirkham 8
*MRC International Nutrition Group, London
School of Hygiene & Tropical Medicine, London,
UK, 2 Muhimbili Wellcome Programme, Muhim-
bili University of Health & Allied Sciences, Dar
es Salaam, Tanzania, 3 Nuffield Department of
Clinical Medicine, University of Oxford, Oxford,
4 Faculty of Medicine, University of Southampton,
Southampton, 5 Department of Psychiatry, Uni-
versity of Oxford, Oxford, UK, 6 Kenya Medical
Research Institute (KEMRI) Centre for Geo-
graphic Medicine Research (Coast), Kilifi, Kenya,
7 Neurological Physics Group, Department of
Medical Physics and Bioengineering, University
Hospital Southampton NHS Foundation Trust,
Southampton, and s Institute of Child Health,
University College London, London, UK
Received 24 October 2013; accepted for
publication 5 January 2014
Correspondence: Dr Sharon Cox, Muhimbili
Wellcome Programme, Muhimbili University
of Health and Allied Sciences, PO Box 65001,
Dar-es-Salaam, Tanzania.
E-mail: Sharon.cox@muhimbili-wellcome.org
Summary
Transcranial Doppler ultrasonography measures cerebral blood flow veloc-
ity (CBFv) of basal intracranial vessels and is used clinically to detect stroke
risk in children with sickle cell anaemia (SCA). Co-inheritance in SCA of
alpha-thalassaemia and glucose-6-phosphate dehydrogenase (G6PD) poly-
morphisms is reported to associate with high CBFv and/or risk of stroke.
The effect of a common functional polymorphism of haptoglobin (HP) is
unknown. We investigated the effect of co-inheritance of these polymor-
phisms on CBFv in 601 stroke-free Tanzanian SCA patients aged <24 years.
Homozygosity for alpha-thalassaemia 3-7 deletion was significantly associ-
ated with reduced mean CBFv compared to wild-type (p-coefficient
— 16-1 cm/s, P = 0 002) adjusted for age and survey year. Inheritance of 1
or 2 alpha-thalassaemia deletions was associated with decreased risk of
abnormally high CBFv, compared to published data from Kenyan healthy
control children (Relative risk ratio [RRR] = 0-53 [95% confidence interval
(CI):0-35-0-8] & RRR = 0-43 [95% CI:0-23-0-78]), and reduced risk of
abnormally low CBFv for 1 deletion only (RRR = 0-38 [95% CI:0T7-
0-83]). No effects were observed for G6PD or HP polymorphisms. This is
the first report of the effects of co-inheritance of common polymorphisms,
including the HP polymorphism, on CBFv in SCA patients resident in
Africa and confirms the importance of alpha-thalassaemia in reducing risk
of abnormal CBFv.
Keywords: sickle cell disease, Africa, children, cerebral blood flow velocity.
Although sickle cell anaemia (SCA) is a monogenic disorder
caused by the homozygous inheritance of sickle haemoglobin
(HbS) resulting in haemolytic anaemia, there is a wide varia-
tion in the severity and pattern of morbidities (Beutler,
2001). The contributions of other co-inherited genetic vari-
ants on risk of severe outcomes such as stroke are under
investigation (Sebastiani et al, 2005; Flanagan et al, 2011).
Stroke is reported to occur in 11% of patients aged <20 years
(Ohene-Frempong et al, 1998; Pandey & Gorelick, 2005).
Transcranial Doppler ultrasonography (TCD) is a well-estab-
lished measure of cerebral blood flow velocity (CBFv) of
basal intracranial vessels and abnormal CBFv may be second-
ary to increased cerebral blood flow or to decrease in diame-
ter of the vessel secondary to vasospasm or stenosis. The
technique is used clinically to detect SCA patients with a
high risk of stroke (Adams et al, 1992) with velocities of
>200 cm/s and >170 cm/s predicting 40% and 7% risk of
stroke over the subsequent 3 years without treatment
© 2014 The Authors. British Journal of Haematology Published by John Wiley & Sons Ltd. First published online 21 February 2014
British Journal of Haematology, 2014, 165, 699-706 doi:1 0.1 1 1 1/bjh. 12791
This is an open access article under the terms of the Creative Commons Attribution License, which permits
use, distribution and reproduction in any medium, provided the original work is properly cited.
BRITISH JOURNAL
OF HAEMATOLOGY
S. E. Cox ef al
(Adams et al, 1992). The risk is substantially reduced by
chronic blood transfusion (Adams et al, 1998). Elevated
CBFv is also associated with increased risk of other neurolog-
ical complications, including seizures (Prengler et al, 2005),
neurocognitive deficits (Krai et al, 2003; Strouse et al, 2006)
and silent infarcts (Pegelow et al, 2001). Low velocities may
also be seen in patients with cerebrovascular disease, consis-
tent with proximal vascular stenosis (Kirkham et al, 1986) or
occlusion (Lee et al, 2004) and appears to be associated with
cerebrovascular accidents (Buchanan et al, 2012).
Haptoglobin (HP) is an acute phase protein that removes
free haemoglobin (Hb) from the circulation after haemolysis
or tissue damage, and is therefore an important anti-oxidant.
HP is encoded by the co-dominant alleles, HP*1 and HP*2,
resulting in three distinct phenotypes HP11, HP12 and
HP22. Plasma concentrations of HP, and its binding affinities
for free Hb, vary by phenotype, with HP22 reported to have
reduced overall Hb affinity (Langlois & Delanghe, 1996), less
efficient antioxidative capacity, (Langlois & Delanghe, 1996;
Melamed-Frank et al, 2001) and reduced clearance rates of
HP2-Hb complexes by macrophages (Asleh et al, 2003). This
is suggested to result in increased nitric oxide scavenging by
HP2-Hb complexes (Azarov et al, 2008), with reduced down-
stream anti-inflammatory signalling after macrophage endo-
cytosis of HP2-Hb complexes (Landis et al, 2013), weaker
inhibition of prostaglandin synthesis and a stronger angio-
genic effect (Cid et al, 1993). There is strong evidence that
the HP22 phenotype is associated with increased risk of
stroke and cardiovascular events in diabetic patients (Vardi
et al, 2012). Furthermore, several reports demonstrate an
association between the HP*2allele with vasospasm after
subarachnoid haemorrhage; as determined by high CBFv
(Borsody et al, 2006), angiographically (Ohnishi et al, 2013),
and in a mouse model (Chaichana et al, 2007). Vasospasm
after sub-arachnoid haemorrhage is thought to result from
inflammation, oxidative stress, nitric oxide scavenging, leuco-
cyte migration and endothelial-leucocyte interactions (Chai-
chana et al, 2010), all from extracellular Hb. In HP22 mice
with sub-arachnoid haemorrhage, treatment with an anti-oxi-
dant (Froehler et al, 2010) or controlled nitric oxide (Momin
et al, 2009) reduces vasospasm, which is supportive evidence
that these mechanisms are important. The HP*2allele is also
associated with increased severity and adverse outcomes in
various infections including human immunodeficiency virus
(HIV and malaria (Cox et al, 2007; McDermid & Prentice,
2006). Effects of HP phenotype on cerebrovascular outcomes
in non-diabetic patients, e.g. those with SCA, have not been
reported.
Although the co-inheritance of oc-thalassaemia deletions
and glucose-6-phosphate dehydrogenase (G6PD) deficiency
have been previously investigated as disease modifiers in
SCA, specifically affecting risk of elevated CBFv and/or
stroke, these have not been investigated in Africa. The
co-inheritance of alpha-thalassaemia in SCA modifies red cell
indices (Embury et al, 1984; Stevens et al, 1986; Kulozik
et al, 1988) and red cell rheology (Serjeant et al, 1983), and,
in some reports, increases total Hb (Embury et al, 1982).
Alpha-thalassaemia is consistently associated with a decreased
risk of increased CBFv and/or stroke in SCA (Adams et al,
1994; Hsu et al, 2003; Bernaudin et al, 2008; Belisario et al,
2010; Flanagan et al, 2011). G6PD deficiency results in
decreased capacity to reduce oxidized glutathione via NADPH
and thus reduced ability of red cells to counteract oxidant
stress (Mason et al, 2007). G6PD deficiency (A-genotype) in
SCA is associated with lower Hb but not increased haemolysis
(Nouraie et al, 2010). Reports of the effects of co- inheritance
of G6PD deficiency with CBFv, vasculopathy on magnetic
resonance angiography and/or stroke risk are contradictory
(Bernaudin et al, 2008, 2011; Rees et al, 2009; Miller et al,
2011; Thangarajh et al, 2012) and there are no data from
patients resident in Africa.
We hypothesized that the co-inheritance of the HP*2allele
in children with SCA would increase the risk of high CBFv
and furthermore, that there may be epistatic effects between
the three unrelated polymorphisms under investigation such
that an effect may be greatest in, or limited to, children with
the HP*2allele who also co-inherited G6PD deficiency and
did not have alpha thalassaemia.
Methods
Ethical approval was granted by the Muhimbili University of
Health and Allied Sciences, Tanzania (MU/RP/AEC/VOL XI/
33), and the London School of Hygiene and Tropical Medi-
cine, UK (reference 5158) Review Boards. All participants,
parents or guardians gave written informed consent (in
Kiswahili) for participation at enrolment into the SCA clini-
cal cohort at Muhimbili National Hospital (MNH).
Study population and Transcranial Doppler examination
Participants aged <24 years were enrolled in the Muhimbili
Sickle Cohort at Muhimbili National Hospital, Tanzania
and had a TCD examination at one or more of three cross-
sectional surveys, conducted in 2004/05, 2009 and 2010.
Patients who received blood transfusion within the previous
two months, symptoms of sickle crisis within the previous
2 weeks or a previous history of stroke were excluded. TCD
examinations were performed according to the STOP
(Stroke prevention in sickle cell disease) protocol (Nichols
et al, 2001) using the Companion II (Nicolet, Warwick,
UK). CBFv was determined as the highest time averaged
maximum velocity in the distal internal carotid artery and
middle cerebral artery on either side. CBFv was classified
using data from 115 Kenyan non-SCA control children
(Newton et al, 1996) as low (<43 or <39-5 cm/s) for the
left and right middle cerebral artery (LMCA/RMCA), high
(>141 or >143-5 cm/s LMCA/RMCA) or normal. CBFv
were also classified as conditional (170-199 cm/s) or abnor-
mal (>200 cm/s) based on previous criteria (Adams et al,
700
© 2014 The Authors. British Journal of Haematology Published by John Wiley & Sons Ltd.
British Journal of Haematology, 2014, 165, 699-706
Polymorphisms & Transcranial Doppler in SCA
1990; Newton et al, 1996). In addition, the difference
between the maximum velocities in the ipsilateral to contra-
lateral MCA was calculated; this was previously shown to
predict vasospasm in patients after sub-arachnoid haemor-
rhage (Nakae et al, 2011). The difference between the two
sides was then calculated as a percentage of the lower
velocity and categorized to indicate the degree of asymme-
try between the left and right MCA as: no asymmetry -
values less than 40%, degree 1; values 40-75%, degree 2;
75-180%, degree 3; and, >180%, degree 4. These cut-off
points are currently used clinically to follow vasospasm
after subarachnoid haemorrhage at University Hospital
Southampton.
Laboratory procedures
Sickle status was diagnosed and quantification of Hb frac-
tions was performed by high performance liquid chromatog-
raphy (HPLC) using the /?-thalassaemia Short Program on
the Variant I analyser (BioRad, Hercules, CA, USA). Full
blood counts were performed using an automated cell coun-
ter (Pentra 60, Horiba ABX, Kyoto, Japan) within 7 d of the
TCD study. Genomic DNA was isolated from peripheral
blood leucocytes using Nucleon kits (BACCII). HP functional
variants (alleles HP*1 and HP*2) were genotyped by allele-
specific polymerase chain reaction (PCR) adapted from
published techniques (Koch et al, 2002; Cox et al, 2007).
Individuals were genotyped for the 3-7 alpha-thalassaemia
deletion using a PCR-based method and agarose gel visuali-
zation as per published methods (Williams et al, 2005). The
202- and 376-single nucleotide polymorphisms (SNPs)
(rsl050828 [G-202A] & rsl050829 [A-376G]), the combined
inheritance of which results in the A- phenotype of glucose
6-phosphate dehydrogenase (G6PD) deficiency, and HbS
(rs334) were determined using multiplex Sequenom® Mass-
ARRAY® (Sequenom®, Hamburg, Germany).
Statistical analysis and sample size
In the current analysis one observation per subject was
selected according to the observation that recorded the high-
est CBFv, assessed separately in the l- and RMCA. Statistical
analyses were performed with Stata IC software (version
12.0, StataCorp, College Station, TX, USA). Categorical vari-
ables were constructed for the CBFv values, using the Kenyan
and STOP cut-off points as described above. We used two
approaches to analyse the CBFv outcome. In the first we
used multiple linear regression to explore predictors of CBFv
as a continuous variable, whilst excluding those with absent
signal. We determined that, using this approach, 90 children
per group (HP22 vs. HP11) were required to detect a differ-
ence in mean CBFv equivalent to 0-5 x standard deviation
(SD) of mean CBFv, based on data from similar aged Kenyan
children with SCA (Makani et al, 2009) with 90% power and
two-tailed significance at 5%. In the second approach, we
used multinomial logistic regression to explore predictors of
CBFv as a categorical variable representing potential cardio-
vascular disease (CVD) (allowing simultaneous investigation
of factors affecting risk of low vs. normal and high vs. low
CBFv, presenting the results as relative risk ratios (RRRs)
with 95% confidence interval (CI). A P-value <005 was
considered statistically significant.
Results
Characteristics of the study population
A total of 601 homozygous SCA patients were included. The
demographic, laboratory and CBFv characteristics of the
patients are summarized in Table I. The age of the patients
ranged from 0-6 to 22-6 years, with 12% aged <5 years and
2-5% aged >16 years. The majority of subjects (51%) were
included from the largest survey in 2004/2005. There was no
evidence of a difference between the surveys for sex, geno-
type or Hb concentrations. However, mean age and HbF%
were significantly lower in the 2004/05 survey. All the partici-
pants had genotype results for at least one of the three poly-
morphisms under investigation, with 385 having complete
data for the three genes being investigated.
Mean CBFv was 131 cm/s (SD = 42). The proportion of
non-detectable CBFv in the LMCA compared to the RMCA
was significantly different (P < 0-006) at 5% for the LMCA
and 8% for the RMCA. However, there was no significant
difference in mean CBFv between the left and right MCA
(120 cm/s [SD 42, N = 571] vs. 119 cm/s [SD 42, N = 552]).
The genotype prevalence data are summarized in Table II.
The prevalence of the inherited combinations of the three
genes in the 385 participants with a complete dataset is avail-
able as supplementary material in Table SI.
Factors associated with TCD outcomes
In the first approach we used linear regression to investigate
possible associations with CBFv for age, sex, Hb and HbF%,
as well as the three genes under investigation, including all
measureable CBFv observations. The results are summarized
in Table III. Strong inverse associations with CBFv were
apparent for age and Hb. Initial analyses of the genotypes
were adjusted for age and year of survey. Hb concentration
was not immediately adjusted for, as all of the genotypes
under investigation may potentially act, at least in part,
through effects on Hb. Inheritance of two copies of the 3-7
cx-thalassaemia deletion, compared to no copies, was signifi-
cantly associated with decreased mean CBFv. If Hb was
adjusted for, no effect of ot-thalassaemia was observed, sug-
gesting the effect is mediated via Hb. There were no apparent
effects of G6PD status or HP genotype, even when when
limited to those with gene combinations hypothesized as the
'worst' and 'best combinations ([HP22/A-A-or A-Z, and/or,
aa/aa] vs. [HPll/BB/-a/-a]),
© 2014 The Authors. British Journal of Haematology Published by John Wiley & Sons Ltd.
British Journal of Haematology, 2014, 165, 699-706
701
S. E. Cox ef al
Table I. Patient characteristics.
Summary
Observations
Demographic variables
statistic
(N)
Age [years], mean (SD)
9-76 (3-86)
601
Males, n (%)
325 (54-08)
601
Survey, n (%)
2004/05
305 (50-75)
601
2009
224 (37-27)
2010
72 (11-98)
Laboratory variables*
Haemoglobin [g/1], mean (SD)
74-0 (11-2)
583
Fetal haemoglobin [%], mean (SD)
5-23 (3-84)
533
CBFv
MCA maximum velocity, left or right,
n (%)
Kenyan non-SCA criteriaf
601
Normal
387 (64-39)
Low [<43 cm/s or <39-5 cm/s]
44 (7-32)
High [>141 cm/s or >143-5 cm/s]
170 (28-29)
STOP criteria
Normal
534 (88-85)
vjOnanionai ^i/u v zuu cm/s
1^ (A 1
ZD ( I - 10)
Abnormal >200 cm/s
42 (6-99)
Asymmetry (Vasospasm):]: n (%)
No asymmetry
317 (54-75)
579
Mild asymmetry
104 (17-96)
Moderate asymmetry
93 (16 06)
Severe asymmetry
65 (11-23)
SD, standard deviation; CBFv, cerebral blood flow velocity; SCA,
sickle cell anaemia; STOP, stroke prevention in sickle cell disease;
MCA: middle cerebral artery.
*A11 laboratory values were assessed within 7 d of the Transcranial
Doppler (TCD) measurement, except HbF which if measured under
the age of 60 months and not within 7 d of the TCD was not
included.
fBased on values from healthy Kenyan children (mean ± 2 SD)
(Newton et al, 1996).
^Degree of asymmetry is difference in the maximum velocities in the
left MCA and right MCA, calculated as a percentage of the lower
velocity. Mild = 40-75%, moderate - 75-180%, severe >180%.
In the second approach we explored associations using mul-
tinomial logistic regression, in which CBFv was categorized as
low, normal or high (Table IV). The only genotype with evi-
dence of an association was the 3-7 a-thalassaemia deletion,
the inheritance of which was associated with a significantly
decreased risk of having an abnormally high or low CBFv.
We also investigated genotype associations with abnormal
CBFv according to the STOP classifications (normal
Table HI. Associations between demographic, laboratory variables
and genetic variants with CBFv as a continuous measurement*.
Observations
Predictors (3
-Coefficient (95% CI)
P-value
(n)
^cr- \y J 1
-2-28 (•
-3-21 to -1-35) <0 0001
525
Qpv (malp vq
3-512 (-
-3-56-10-58)
0-329
525
female) t
Haemoglobin
-4-60 (
-7-91 to-1-29)
0007
525
(g/Dt
Fetal
0-197 (
-1-11-0-72)
0-672
525
haemoglobin
(%)t
Haptoglobin HP lit
HP12
0-28 (■
-9-61-10-17)
0-956
410
HP22
■2-994 (■
-15-32-9-33)
0-633
ot-Thalassaemia normal^
1 Deletion
-6-32 (
-13-78-1-14)
0097
549
2 Deletions -
1614 (
-26-271
0002
to -6 0135)
G6PD Phenotype normal^
Mild (A-B)
-8-80 (■
-19-44-1-84)
0105
564
Affected (A-Z
-1-81 (
-11-83-8-21)
0-723
or A-A-)
*Only velocity values >0 cm/s included for multiple linear regression.
fResults for multivariable regression including the marked variables.
^Adjusted for age and survey year CBFv, cerebral blood flow velocity;
95% CI, 95% confidence interval.
<170 cm/s, conditional 10-199 cm/s and abnormal >200 cm/
s) (Adams et al, 1998) but no significant associations were
observed.
Finally we also investigated possible associations with the
degree of asymmetry in CBFv, classified as normal, mild,
moderate or severe. No consistent effects were observed for
any of the variables tested (Table V).
Discussion
This is the largest study to date to assess the effects of the
disease modifying and commonly co-inherited polymor-
phisms of alpha-thalassaemia and G6PD on prospectively
measured CBFv in SCA, and the first in children and adoles-
cents resident in Africa. In addition, this is the first study to
assess the effect of co-inheritance of HP polymorphisms on a
clinical end-point in SCA. We confirm previous reports of a
reduction in mean CBFv and protection from abnormal
CBFv measurements in heterozygote and homozygotes for
Table II. Genotype prevalence and scores assigned to each genotype.
Gene
Homozygote WT
Heterozygote
Homo/hemizygote mutant
G6PD A-Phenotype (202 & 376) N = 583
BB [75-30%]
A-B [11-32%]
A-Z or A-A-[13-38%]
3-7 Alpha-thalassaemia deletion N = 568
aalaa [43-66%]
tx-/acx [40-67%]
(X-/<X- [15-67%]
HPN = 422
HP11 [23-46%]
HP12 [56-64%]
HP22 [19-91%]
702
© 2014 The Authors. British Journal of Haematology Published by lohn Wiley & Sons Ltd.
British Journal of Haematology, 2014, 165, 699-706
Polymorphisms & Transcranial Doppler in SCA
Table IV. Relative risk ratios of genetic variants on CBFv categories
defined as low or high compared to normal using data from Kenyan
healthy non-SCA child population adjusted for age and year of
survey.
Relative risk ratio
(95% confidence interval)*
Genotypes
(CBFv Low
vs. normal)
(CBFv High
vs. normal)
iV
Haptoglobin, HP 11
HP12 0-90 (0-33-2-47)
HP22 1 01 (0-30-3-38)
oi-Thalassaemia, normal
1 Deletion 0-38 (0-17-0-83)f
2 Deletions 0-95 (0-41-2-20)
G6PD Normal phenotype
Mild 0-49 (0-14-1-65)
Affected 0-41 (1-12-1-37)
90 (0-53-1-51)
85 (0-44-1-66)
53 (0-35-0-80) +
43 (0-23-0-78)f
62 (0-32-1-17)
78 (0-44-1-39)
422
568
583
*Low CBFv corresponds to values <43 in LMCA or <39-5 in RMCA,
including values of 0; High CBFv corresponds to values >141-1 in
LMCA or >143-6 in RMCA), compared to normal (CBFv: 43-
141 cm/s in LMCA or 39-6-143-5 cm/s in RMCA) (Newton et al,
1996).
tP-value <0 05.
RRR: Relative risk ratio; CBFv, cerebral blood flow velocity; SCA,
sickle cell anaemia; RMCA, right middle cerebral artery; LMCA, left
middle cerebral artery.
the oe-thalassaemia 3-7 deletion. This effect was not signifi-
cant when limited to abnormal, as defined by the STOP
criteria. We also observed a significantly decreased risk of
abnormally low CBFv in alpha thalassaemia heterozygote
children. No effects of G6PD or HP polymorphisms were
observed, including when assessed in children with the
hypothesized 'worst' gene combinations compared to 'opti-
mal' combination.
Mean CBFv (131 cm/s) and the prevalence of highly
elevated (>200 cm/s) (7%) and conditionally elevated (170-
199 cm/s) (4%) CBFv were higher in our Tanzanian patients
compared to rural Kenyan children with SCA with a mean
CBFv of 120 cm/s and only 3% with conditional and none
with high CBFv (Makani et al, 2009), but still considerably
lower than in other studies in which alpha-thalassaemia and/
or G6PD were also assessed. Sixteen percent of French chil-
dren with sickle cell disease had CBFv >200 cm/s (62/373)
(Bernaudin et al, 2008), whilst 18% (409/2334) had CBFv of
170-199 cm/s and 9% >200 cm/s (217/2324) in the baseline
measurement of the STOP study in the USA (Adams et al,
2004). However, the proportions were similar to that
reported in Brazilian patients with SCA/sickle cell disease in
whom 4% (7/164) had CBFv 170-199 cm/s and 3% (5/164)
had a CBFv >200 m/s (Belisario et al, 2010). In both the
Brazilian and the French cohorts the 3-7 alpha-thalassaemia
deletion was associated with protection from conditional or
high CBFv, whilst in the STOP trial cohort a similar finding
was observed in a case control design (Hsu et al, 2003). Thus
our results confirm a protective effect of the alpha-thalassae-
mia 3-7 deletion, resulting in a lower mean CBFv and reduc-
ing the risk of even moderate elevations, compared to
normal, and importantly, also protecting against abnormally
low CBFv, also an indicator of CVD and stroke risk. How-
ever, the overall protective effect of alpha-thalassaemia status
on risk of stroke in our patients remains to be determined
and may differ from other populations less dominated by the
Central African Republic (CAR) haplotype. No protective
Table V. Relative risk ratios of
genetic variants on category
of vascular asymmetry as an indicator
of vasospasm adjusted for a;
;e and year of
survey.
Relative risk ratio (95% confidence interval)*
Genotypes
Mild vs. normal
Moderate vs. normal
Severe vs. normal
N
Aget
0-08 (0-02-0-14) +
-001 (-0-08-0-05)
0 03 (-0 04-0 11)
525
Sext
0-01 (-0-48-0-50)
-0-08 (-0-58-0-42)
0-04 (-0-54-0-62)
Fetal Haemoglobinf
-0 03 (-0-09-0-04)
-0 02 (-0-09-0-04)
-0 03 (-0 11-0 05)
Haemoglobinf
010 (-0-13-0-34)
-0-27 (-0-50 to -0-03)+.
-0-08 (-0-35-0-20)
Haptoglobin, HP11
HP12
1-07 (0-58-1-98)
1-12 (0-56-2-21)
1-25 (0-56-2-79)
410
HP22
0-86 (0-39-1-90)
1-21 (0-53-2-75)
1-12 (0-41-3-03)
cx-Thalassaemia, normal
1 Deletion
1-69 (1-01-2-84) +
0-98 (0-58-1-64)
0-94 (0-52-1-70)
549
2 Deletions
1-66 (0-86-3-23)
0-99 (0-49-1-98)
0-52 (0-20-1-35)
G6PD Normal phenotype
Mild (A-B)
0-98 (0-48-1-98)
0-94 (0-44-1-99)
0-95 (0-39-2-31)
564
Affected (A-Z or A-A-)
0-77 (0-38-1-58)
1 11 (0-57-2-16)
0-85 (0-35-2-03)
*Degree of asymmerty is difference in the maximum velocities in LMCA and RMCA calculated as a percentage of the lower velocity. Mild
75%, moderate -75-180%, severe >180%.
f Results for multivariable regression including the marked variables.
JP-value <0 05; Adjusted for survey year.
40-
© 2014 The Authors. British Journal of Haematology Published by John Wiley & Sons Ltd.
British Journal of Haematoiogy, 2014, 165, 699-706
703
S. E. Cox ef al
effect of altha-thalassaemia was apparent for degree of asym-
metry between the ipsilateral and contralateral MCA. The
effect of the alpha-thalassaemia deletion may be mediated at
least in part via increased Hb. Co-inheritance of the 3-7 dele-
tion is consistently and quantitatively associated with
decreased mean cell volume (MCV), mean cell haemoglobin
(MCH), mean cell haemoglobin concentration (MCHC) and
haemolytic markers, including in this population, (Cox et al,
2013) and with decreased WBC (Bernaudin et al, 2008). HbS
concentration is the primary determinant of HbS polymeri-
zation (Eaton & Hofrichter, 1987). Thus the protective effect
of co-inheritance of the 3-7 deletion has been commonly
assumed to be due to decreased sickling rates (Ballas, 2001)
and consequent haemolysis. However, in the analysis by
Bernaudin et al (2008) there were independent effects on
CBFv of the 3-7 deletion after adjustment for MCV (MCH &
MCHC not analysed), white blood cell count and the haemo-
lytic marker lactate dehydrogenase (LDH) (itself indepen-
dently associated in the multivariate model). The authors
suggested increased red cell deformability as a potential
mechanism. In a recent, large, carefully controlled study, co-
inheritance of the 3-7 deletion in SCA was quantitatively and
negatively associated with the proportion of dense dehy-
drated red cells (%DRBCs) measured at steady state.
Increased%DRBC was, in turn, positively associated with risk
of priapism, leg ulcers and renal dysfunction, of which only
renal dysfunction was associated with haemolysis (Bartolucci
et al, 2012). There is evidence to suggest that the increased
rigidity of DBRCs may promote vasoconstriction via reduced
shear stress-induced ATP release as a vasodilatory signalling
molecule (Wan et al, 2008).
It is unlikely that the lack of an effect of the HP*2allele
was due to a type II error as a result of a lack of power, as
evidenced by our adequate sample size of 83 HP22 and 103
HP 11 compared to our sample size estimate of 90 per
group to detect an effect equivalent to half a standard devi-
ation (SD) in CBFv. Thus although there was a probable
increased variance in our data due to different observers
between survey years, a similar effect size was observed for
alpha-thalassaemia in this study (oc-/ot- (JV = 94) vs. ora/cta
(N = 258), P-coefficient -16-14 cm/s = 0-4 SD of CBFv).
However, the 95% CIs of the observed effect of HP22 did
just include the effect size in the sample size calculation
(—15-3 cm/s vs. 0-5 x SD = 20-9). Thus, it remains a possi-
bility that small effects of the HP*2allele exist, reducing risk
of either both or the lower extremes of CBFv, not captured
in the multinomial logistic model, which was not included
in pre-study sample size calculations. Thus despite the
strong evidence for a mechanism whereby the HP22 variant
could be expected to increase vascular activation and vaso-
constriction in the inflammatory and haemolytic condition
of SCA, our data does not support this suggestion. An
explanation for this could be that HP is overwhelmed in
SCA, such that phenotypic variations in its functions cease
to be relevant. This scenario is supported by our observation
that during well, steady-state clinic visits, HP levels were
non-detectable in 96% of our patients (lower limit of
detection = 0-04 g/1, N = 838) compared to 17% of non-
SCA controls and not affected by genotype (S. E. Cox, J.
Makani & A. M. Prentice, unpublished data). A possible
alternative explanation is that in HP22 individuals the haem-
oxygenase-1 compensatory pathway may be more successfully
up-regulated in response to haem exposure, demonstrated to
have powerful anti-inflammatory and anti-stasis effects (Jison
et al, 2004; Belcher et al, 2006).
Conflicting reports of an effect of G6PD on CBFv (Ber-
naudin et al, 2008, 2011; Rees et al, 2009; Miller et al, 2011;
Thangarajh et al, 2012) may result from variations between
populations in either the phenotypic expression compared to
assessed genotype or from population or methodological dif-
ferences between in-vivo vs. in-vitro enzyme activity, when
this has been measured directly (Johnson et al, 2009). There
is little evidence for an association of either low enzyme
activity (Miller et al, 2011) or genotype (Flanagan et al,
2011) with stroke. Decreased enzyme activity was associated
with TCD abnormality in children resident in France
(Bernaudin et al, 2008) and this effect appeared to be inde-
pendent of haemolysis. However, this association was not
observed in England (Rees et al, 2009). Genotypes rsl050828
or rsl050829 were associated with magnetic resonance angi-
ography evidence of vasculopathy in a cohort from the US,
England and France but children with TCD abnormality were
excluded (Thangarajh et al, 2012). It is possible that G6PD
polymorphisms play a role in the initiation or development
of vasculopathy in certain environments but we have no
evidence for this in our population resident in Africa and the
lack of an association with stroke suggests that other risk
factors may be more important.
In conclusion, despite adequate power to determine a clin-
ically relevant effect size of the _HP*2allele on CBFv, we could
determine no evidence of such an effect. In a population of
SCA patients resident in Africa, we confirmed the effect of
alpha-thalassaemia, but could not find an effect of G6PD.
Acknowledgements
We warmly thank the patients and staff of MNH and
MUHAS, Dar-es-Salaam, Tanzania who made this work
possible. We also thank Gurishaeli Walter and Abdulnar
Maupa for TCD measurements in 2009, Alex Macharia and
Rachel Wiltshire for assistance with the genotyping, Jose-
phine Mgaya and Harvest Mariki for haemoglobin quantifi-
cation by HPLC, Lazarus Mramba and Anthony Fulford for
statistical discussions and Tom Williams for his review of the
manuscript. SEC, JM, FJK, CRJN 8c AMP designed the
research. JM, DS, VSE, EK & AAB performed the research.
AAB contributed essential analysis tools and data. SEC, FJK
8c VSE analysed the data. SEC 8c FJK wrote the paper. All
authors contributed to drafting and/or critical revisions of
the manuscript and approved the final version.
704
© 2014 The Authors. British Journal of Haematology Published by John Wiley & Sons Ltd.
British Journal of Haematology, 2014, 165, 699-706
Polymorphisms & Transcranial Doppler in SCA
Funding
This work was supported by the Wellcome Trust UK: (SC)
project grant 080025; (JM) personal fellowship 072064; stra-
tegic awards 084538 and 095009.
Supporting Information
Additional Supporting Information may be found in the
online version of this article:
Table SI. Proportions of subjects for each combination of
the three genotypes under investigation.
References
Adams, R.J., Nichols, F.T. 3rd, Aaslid, R., McKie,
V.C., McKie, K., Carl, E., Stephens, S., Thomp-
son, W.O., Milner, P. & Figueroa, R. (1990)
Cerebral vessel stenosis in sickle cell disease:
criteria for detection by transcranial Doppler.
The American Journal of Pediatric Hematology/
Oncology, 12, 277-282.
Adams, R.J., Nichols, F.T., Figueroa, R., McKie, V.
& Lott, T. (1992) Transcranial Doppler correla-
tion with cerebral angiography in sickle cell dis-
ease. Stroke, 23, 1073-1077.
Adams, R.J., Kutlar, A., McKie, V., Carl, E., Nic-
hols, F.T., Liu, J.C., McKie, K. & Clary, A.
(1994} Alpha thalassemia and stroke risk in
sickle cell anemia. American Journal of Hematol-
ogy, 45, 279-282.
Adams, R.J., McKie, V.C., Hsu, L., Files, B.,
Vichinsky, E., Pegelow, C, Abboud, M., Galla-
gher, D., Kutlar, A., Nichols, F.T., Bonds, D.R.
& Brambilla, D. (1998) Prevention of a first
stroke by transfusions in children with sickle cell
anemia and abnormal results on transcranial
Doppler ultrasonography. New England Journal
of Medicine, 339, 5-11.
Adams, R.J., Brambilla, D.J., Granger, S., Galla-
gher, D., Vichinsky, E., Abboud, M.R., Pegelow,
C.H., Woods, C, Rohde, E.M., Nichols, F.T.,
Jones, A., Luden, J.P., Bowman, L., Hagner, S.,
Morales, K.H. & Roach, E.S. (2004) Stroke and
conversion to high risk in children screened
with transcranial Doppler ultrasound during the
STOP study. Blood, 103, 3689-3694.
Asleh, R., Marsh, S., Shilkrut, M., Binah, O., Guet-
ta, J., Lejbkowicz, F., Enav, B., Shehadeh, N.,
Kanter, Y., Lache, O., Cohen, O., Levy, N.S. &
Levy, A.P. (2003) Genetically determined hetero-
geneity in hemoglobin scavenging and suscepti-
bility to diabetic cardiovascular disease.
Circulation Research, 92, 1193-1200.
Azarov, I., He, X., Jeffers, A., Basu, S., Ucer, B.,
Hantgan, R.R., Levy, A. & Kim-Shapiro, D.B.
(2008) Rate of nitric oxide scavenging by hemo-
globin bound to haptoglobin. Nitric Oxide, 18,
296-302.
Ballas, S.K. (2001) Effect of alpha-globin genotype
on the pathophysiology of sickle cell disease.
Pediatric Pathology & Molecular Medicine, 20,
107-121.
Bartolucci, P., Brugnara, C„ Teixeira-Pinto, A.,
Pissard, S., Moradkhani, K., Jouault, H. &
Galacteros, F. (2012) Erythrocyte density in
sickle cell syndromes is associated with specific
clinical manifestations and hemolysis. Blood,
120, 3136-3141.
Belcher, J.D., Mahaseth, H., Welch, T.E., Otter-
bein, L.E., Hebbel, R.P. & Vercellotti, G.M.
(2006} Heme oxygenase- 1 is a modulator of
inflammation and vaso-occlusion in transgenic
sickle mice. Journal of Clinical Investigation, 116,
808-816.
Belisario, A.R., Rodrigues, C.V., Martins, M.L.,
Silva, CM. & Viana, M.B. (2010) Coinheritance
of alpha-thalassemia decreases the risk of cere-
brovascular disease in a cohort of children with
sickle cell anemia. Hemoglobin, 34, 516-529.
Bcrnaudin, F., Verlhac, S., Chevret, S., Torres, M.,
Coic, L., Arnaud, C, Kamdem, A., Hau, I.,
Grazia Neonato, M. & Delacourt, C. (2008)
G6PD deficiency, absence of alpha-thalassemia,
and hemolytic rate at baseline are significant
independent risk factors for abnormally high
cerebral velocities in patients with sickle cell
anemia. Blood, 112, 4314-^317.
Bernaudin, F., Verihac, S., Arnaud, C, Kamdem,
A., Chevret, S., Hau, I., Coic, L., Leveille, E.,
Lemarchand, E., Lesprit, E., Abadie, I., Medejel,
N., Madhi, F., Lemerle, S., Biscardi, S., Bard-
akdjian, J., Galacteros, F., Torres, M., Kuentz,
M., Ferry, C, Socie, G., Reinert, P. & Delacourt,
C. (2011) Impact of early transcranial Doppler
screening and intensive therapy on cerebral vas-
culopathy outcome in a newborn sickle cell ane-
mia cohort. Blood, 117, 1130-1140; quiz 1436.
Beutler, E. (2001} Discrepancies between genotype
and phenotype in hematology: an important
frontier. Blood, 98, 2597-2602.
Borsody, M., Burke, A., Coplin, W., Miller-Lotan,
R. & Levy, A. (2006) Haptoglobin and the
development of cerebral artery vasospasm after
subarachnoid hemorrhage. Neurology, 66, 634-
640.
Buchanan, I.D., James-Herry, A. & Osunkwo, I.
(2012} The other side of abnormal: a case series
of low transcranial Doppler velocities associated
with stroke in children with sickle cell disease.
Jo u rnal of Pedia trie Hem a to logy I o neology, 3 5 ,
543-546.
Chaichana, K.L., Levy, A.P., Miller-Lotan, R.,
Shakur, S. & Tamargo, R.J. (2007) Haptoglobin
2-2 genotype determines chronic vasospasm
after experimental subarachnoid hemorrhage.
Stroke, 38, 3266-3271.
Chaichana, K.L., Pradilla, G., Huang, J. & Tamar-
go, R.J. (2010) Role of inflammation (leukocyte-
endothclial cell interactions) in vasospasm after
subarachnoid hemorrhage. World Neurosurgery,
73, 22-41.
Cid, M.C., Grant, D.S., Hoffman, G.S., Auerbach,
R., Fauci, A.S. & Kleinman, H.K. (1993) Identi-
fication of haptoglobin as an angiogenic factor
in sera from patients with systemic vasculitis.
Journal of Clinical Investigation, 91, 977-985.
Cox, S.E., Doherty, C, Atkinson, S.H., Nweneka,
C.V., Fulford, A.J., Ghattas, H., Rockett, K.A.,
Kwiatkowski, D.P. & Prentice, A.M. (2007)
Haplotype association between haptoglobin
(Hp2) and Hp promoter SNP (A-61C) may
explain previous controversy of haptoglobin and
malaria protection. PLoS One, 2, e362.
Cox, S.E., Makani, J., Newton, C.R., Prentice, A.M.
& Kirkham, F.J. (2013) Hematological and
genetic predictors of daytime hemoglobin satu-
ration in tanzanian children with and without
sickle cell anemia. ISRN Hematology, 2013,
472909.
Eaton, W.A. & Hofrichter, J. (1987) Hemoglobin S
gelation and sickle cell disease. Blood, 70, 1245-
1266.
Embury, S.H., Dozy, A.M., Miller, J., Davis, J.R. Jr,
Kleman, K.M., Preisler, H., Vichinsky, E., Lande,
W.N., Lubin, B.H., Kan, Y.W. & Mentzer, W.C.
(1982} Concurrent sickle-cell anemia and alpha-
thalassemia: effect on severity of anemia. New
England Journal of Medicine, 306, 270-274.
Embury, S.H., Clark, M.R., Monroy, G. & Mohan-
das, N. (1984) Concurrent sickle cell anemia
and alpha-thalassemia. Effect on pathological
properties of sickle erythrocytes. Journal of
Clinical Investigation, 73, 116-123.
Flanagan, J.M., Frohlich, D.M., Howard, T.A.,
Schultz, W.H., Driscoll, C, Nagasubramanian,
R., Mortier, N.A., Kimble, A.C., Aygun, B.,
Adams, R.J., Helms, R.W. & Ware, R.E. (2011)
Genetic predictors for stroke in children with
sickle cell anemia. Blood, 117, 6681-6684.
Froehler, M.T., Kooshkabadi, A., Miller-Lotan, R.,
Blum, S., Sher, S., Levy, A. & Tamargo, R.J.
(2010} Vasospasm after subarachnoid hemor-
rhage in haptoglobin 2-2 mice can be prevented
with a glutathione peroxidase mimetic. Journal
of Clinical Neuroscience, 17, 1169-1172.
Hsu, L.L., Miller, S.T., Wright, E., Kutlar, A.,
McKie, V., Wang, W., Pegelow, C.H., Driscoll,
C, Hurlet, A., Woods, G., Elsas, L., Embury, S.
& Adams, R.J. (2003) Alpha Thalassemia is asso-
ciated with decreased risk of abnormal transcra-
nial Doppler ultrasonography in children with
sickle cell anemia. Journal of Pediatric Hematol-
ogy/ oncology, 25, 622-628.
Jison, M.L., Munson, P.J., Barb, J. J., Suffredini,
A.F., Talwar, S., Logun, C, Raghavachari, N.,
Beigel, J.H., Shelhamer, J.H., Danner, R.L. &
Gladwin, M.T. (2004} Blood mononuclear cell
gene expression profiles characterize the oxidant,
hemolytic, and inflammatory stress of sickle cell
disease. Blood, 104, 270-280.
© 2014 The Authors. British Journal of Haematology Published by John Wiley & Sons Ltd.
British Journal of Haematoiogy, 2014, 165, 699-706
705
S. E. Cox et at
Johnson, M.K., Clark, T.D., Njama-Meya, D.,
Rosenthal, P.J. & Parikh, S. (2009) Impact of
the method of G6PD deficiency assessment on
genetic association studies of malaria susceptibil-
ity. PLoS One, 4, e7246.
Kirkham, F.J., Neville, B.G. & Levin, S.D. (1986)
Bedside diagnosis of stenosis of middle cerebral
artery. Lancet, 1, 797-798.
Koch, W., Latz, W., Eichinger, M., Roguin, A.,
Levy, A.P., Schomig, A. & Kastrati, A. (2002)
Genotyping of the common haptoglobin Hp 1/2
polymorphism based on PCR. Clinical Chemis-
try, 48, 1377-1382.
Krai, M.C., Brown, R.T., Nietert, P.J., Abboud,
M.R., Jackson, S.M. & Hynd, G.W. (2003)
Transcranial Dopplcr ultrasonography and neu-
rocognitive functioning in children with sickle
cell disease. Pediatrics, 112, 324-331.
Kulozik, A.E., Kar, B.C., Serjeant, G.R., Serjeant,
B.E. & Weatherall, D.J. (1988) The molecular
basis of alpha thalassemia in India. Its interac-
tion with the sickle cell gene. Blood, 71,
467-472.
Landis, R.C., Philippidis, P., Domin, J., Boyle, J.J.
& Haskard, D.O. (2013) Haptoglobin genotype-
dependent anti- inflammatory signaling in
CD163(+) macrophages. International Journal of
Inflammation, 2013, 980327.
Langlois, M.R. & Delanghe, J.R. (1996) Biological
and clinical significance of haptoglobin poly-
morphism in humans. Clinical Chemistry, 42,
1589-1600.
Lee, Y.S., Jung, K.H. & Roh, J.K. (2004) Diagnosis
of moyamoya disease with transcranial Doppler
sonography: correlation study with magnetic
resonance angiography. Journal of Neuroimaging,
14, 319-323.
Makani, J., Kirkham, F.J., Komba, A., Ajala-Agbo,
T., Otieno, G., Fegan, G., Williams, T.N., Marsh,
K. & Newton, C.R. (2009) Risk factors for high
cerebral blood flow velocity and death in Ken-
yan children with Sickle Cell Anaemia: role of
haemoglobin oxygen saturation and febrile
illness. British Journal of Haematology, 145,
529-532.
Mason, P. J., Bautista, J.M. & Gilsanz, F. (2007)
G6PD deficiency: the genotype-phenotype
association. Blood Reviews, 21, 267-283.
McDermid, J.M. & Prentice, A.M. (2006) Iron and
infection: effects of host iron status and the
iron-regulatory genes haptoglobin and NRAMP1
(SLC11A1) on host-pathogen interactions in
tuberculosis and HIV. Clinical Science (London),
110, 503-524.
Melamed -Frank, M., Lache, O., Enav, B.I., Szafra-
nek, T., Levy, N.S., Ricklis, R.M. & Levy, A.P.
(2001) Structure-function analysis of the antiox-
idant properties of haptoglobin. Blood, 98,
3693-3698.
Miller, S.T., Milton, J. & Steinberg, M.H. (2011)
G6PD deficiency and stroke in the CSSCD.
American Journal of Hematology, 86, 331.
Momin, E.N., Schwab, K.E., Chaichana, K.L.,
Millcr-Lotan, R., Levy, A.P. & Tamargo, R.J.
(2009) Controlled delivery of nitric oxide inhib-
its leukocyte migration and prevents vasospasm
in haptoglobin 2-2 mice after subarachnoid
hemorrhage. Neurosurgery, 65, 937-945; discus-
sion 945.
Nakae, R., Yokota, H., Yoshida, D. & Teramoto,
A. (2011) Transcranial Dopplcr ultrasonography
for diagnosis of cerebral vasospasm after aneu-
rysmal subarachnoid hemorrhage: mean blood
flow velocity ratio of the ipsilateral and contra-
lateral middle cerebral arteries. Neurosurgery, 69,
876-883; discussion 883.
Newton, C.R., Marsh, K., Peshu, N. & Kirkham,
F.J. (1996) Perturbations of cerebral hemody-
namics in Kenyans with cerebral malaria. Pediat-
ric Neurology, 15, 41-49.
Nichols, F.T., Jones, A.M. & Adams, R.J. (2001)
Stroke prevention in sickle cell disease (STOP)
study guidelines for transcranial Doppler testing.
Journal of Neuroimaging, 11, 354-362.
Nouraie, M., Reading, N.S., Campbell, A., Minniti,
CP., Rana, S.R., Luchtman-Jones, L., Kato, G.J.,
Gladwin, M.T., Castro, O.L., Prchal, J.T. & Gor-
deuk, V.R. (2010) Association of G6PD with lower
haemoglobin concentration but not increased
haemolysis in patients with sickle cell anaemia.
British Journal of Haematology, 150, 218-225.
Ohene-Frempong, K., Weiner, S.J., Sleeper, L.A.,
Miller, S.T., Embury, S., Moohr, J.W., Wethers,
D.L., Pegelow, C.H. & Gill, F.M. (1998) Cere-
brovascular accidents in sickle cell disease: rates
and risk factors. Blood, 91, 288-294.
Ohnishi, H., Iihara, K., Kaku, Y., Yamauchi, K.,
Fukuda, K., Nishimura, K., Nakai, M., Satow,
T., Nakajima, N. & Ikegawa, M. (2013) Hapto-
globin phenotype predicts cerebral vasospasm
and clinical deterioration after aneurysmal sub-
arachnoid hemorrhage. Journal of Stroke and
Cerebrovascular Diseases, 22, 520-526.
Pandey, D.K. & Gorelick, P.B. (2005) Epidemiol-
ogy of stroke in African Americans and His-
panic Americans. Medical Clinics of North
America, 89, 739-752, vii.
Pegelow, C.H., Wang, W., Granger, S., Hsu, L.L.,
Vichinsky, E., Moser, F.G., Bello, J., Zimmer-
man, R.A., Adams, R.J., Brambilla, D. & Trial,
S. (2001) Silent infarcts in children with sickle
cell anemia and abnormal cerebral artery veloc-
ity. Archives of Neurology, 58, 2017-2021.
Prengler, M., Pavlakis, S.G., Boyd, S., Connelly, A.,
Calamante, F., Chong, W.K., Saunders, D., Cox,
T., Bynevelt, M., Lane, R., Laverty, A. & Kirk-
ham, F.J. (2005) Sickle cell disease: ischemia and
seizures. Annals of Neurology, 58, 290-302.
Rees, D.C., Lambert, C, Cooper, E., Bartram, J.,
Goss, D., Deane, C. & Thein, S.L. (2009) Glu-
cose 6 phosphate dehydrogenase deficiency is
not associated with cerebrovascular disease in
children with sickle cell anemia. Blood, 114,
742-743; author reply 743-744.
Sebastiani, P., Ramoni, M.F., Nolan, V., Baldwin,
C.T. & Steinberg, M.H. (2005) Genetic dissec-
tion and prognostic modeling of overt stroke in
sickle cell anemia. Nature Genetics, 37, 435-440.
Serjeant, B.E., Mason, K.P., Kenny, M.W., Stuart,
J., Higgs, D.R., Weatherall, D.J., Hayes, R.J. &
Serjeant, G.R. (1983) Effect of alpha thalassae-
mia on the rheology of homozygous sickle cell
disease. British Journal of Haematology, 55, 479-
486.
Stevens, M.C., Maude, G.H., Beckford, M., Grand-
ison, Y., Mason, K., Taylor, B., Serjeant, B.E.,
Higgs, D.R., Teal, H. & Weatherall, D.J. (1986)
Alpha thalassemia and the hematology of homo-
zygous sickle cell disease in childhood. Blood,
67, 411-414.
Strouse, J.J., Cox, C.S., Melhem, E.R., Lu, H.,
Kraut, M.A., Razumovsky, A., Yohay, K., van
ZijL P.C. & Casella, J.F. (2006) Inverse correla-
tion between cerebral blood flow measured
by continuous arterial spin-labeling (CASL)
MRI and neurocognitive function in children
with sickle cell anemia (SCA). Blood, 108, 379-
381.
Thangarajh, M., Yang, G., Fuchs, D., Ponisio,
M.R., McKinstry, R.C., Jaju, A., Noetzel, M.J.,
Casella, J.F., Barron-Casclla, E., Hooper, W.C.,
Boulet, S.L., Bean, C.J., Pyle, M.E., Payne, A.B.,
Driggers, J., Trau, H.A., Vcndt, B.A., Rodeghier,
M. & DeBaun, M.R. (2012) Magnetic resonance
angiography-defined intracranial vasculopathy is
associated with silent cerebral infarcts and glu-
coses-phosphate dehydrogenase mutation in
children with sickle cell anaemia. British Journal
of Haematology, 159, 352-359.
Vardi, M., Blum, S. & Levy, A.P. (2012) Haptoglo-
bin genotype and cardiovascular outcomes in
diabetes mellitus - natural history of the disease
and the effect of vitamin E treatment. Meta-
analysis of the medical literature. European
Journal of Internal Medicine, 23, 628-632.
Wan, J., Ristenpart, W.D. & Stone, H.A. (2008)
Dynamics of shear-induced ATP release from
red blood cells. Proceedings of the National Acad-
emy of Sciences of the United States of America,
105, 16432-16437.
Williams, T.N., Mwangi, T.W., Wambua, S., Peto,
T.E., Weatherall, D.J., Gupta, S., Recker, M.,
Penman, B.S., Uyoga, S., Macharia, A.,
Mwacharo, J.K., Snow, R.W. & Marsh, K.
(2005) Negative epistasis between the malaria-
protective effects of alpha+ thalassemia and the
sickle cell trait. Nature Genetics, 37, 1253-1257.
706
© 2014 The Authors. British Journal of Haematology Published by John Wiley & Sons Ltd.
British Journal of Haematology, 2014, 165 f 699-706